> We found people who had received the emails directly rated the messages far more negatively than did the observers. I wonder, how did they control for the co

> We found people who had received the emails directly rated the messages far more negatively than did the observers. I wonder, how did they control for the context? Obviously, the person who received the mail knows much more context about it, than a side observer.